Professional Documents
Culture Documents
Berger Reviewed work(s): Source: American Journal of Sociology, Vol. 83, No. 4 (Jan., 1978), pp. 823-861 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/2777718 . Accessed: 13/08/2012 13:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Sociology.
http://www.jstor.org
Coup d'Etat,
Organizational change and conflict are conceptualized in social movement terms. Using an analogy to social movements in the nationstate, a resource-mobilization approach is employed to examine the range of social movement phenomena in organizations. Three major forms are distinguished-the coup d'etat, bureaucratic insurgency, and mass movements-which differ in their breadth, location in the organizational social structure, goals, and tactics. Major attention is given to the occurrence of social movements in corporate hierarchical forms, although the approach is applicable also to social movement phenomena in federated and in voluntary associations. A number of illustrative hypotheses are given. Power structures and political processes of organizations are important components of organizational analysis. Theoretical perspectives as disparate as those of Cyert and March (1963) and Zald (1970a, 1970b) contain notions of political processes or political structure as central components. And a large number of empirical studies examine the role of politics and power in the allocation of resources (Pfeffer and Salancik 1974) as determinants of organizational satisfaction (Tannenbaum 1962); as factors in the evolution of procedural rights, due process, and quasi-judicial processes (Evan 1965; Pondy 1964); and as influencing decisions on major capital investments or acquisitions such as computers (Pettigrew 1973). Moreover, a number of papers treat power distributions as the outcomes of organizational tasks, contingencies, or dependencies (Crozier 1964; Thompson 1967; Perrow 1970; Zald 1962; and Hickson et al. 1971). Although this literature is rich and growing, it has two noticeable shortcomings. First, because so many of the organizational studies are synchronic, the study of things political is only rarely linked (either as independent or
are indebted to Ivar Berg, Anthony Oberschall, David Snyder, and the reviewers of this Joutrnalfor many of the illustrations used in this paper and for discussions of the issues raised. An earlier version was presented at the meeting of the American Sociological Association, New York, September 1976. The Vanrderbilt University Research Council has aided in the preparation of the manuscript. 1978 by The University of Chicago. All rights reserved. 06D 0002 -9602/78/8304-0001$02 .98
1 We
823
American Journal of Sociology as dependent variables) to the study of organizational change. Second, studies or theories of organizationalpolitics rarely take note of their kinship with investigations of political processes in nations and communities (but see Zald 1965). Thus they ignore a massive amount of thinking and research that might be relevant to work on organizations. The fundamental assumption of this essay is that social movements or phenomena resembling them occur in organizations.That is to say, just as political sociologists study social movements in nation states because they represent the cause and effect of societal change occurring outside institutionalized channels, so, we believe, the student of organizationsneeds to examine social movement phenomena in organizations because they too may be the cause and effect of emergingorganizationalchange. Implied by our assumption is the use of social movements in the larger society as models of or analogies to activities within organizations. For example, in the fall of 1975 Robert Sarnoff, president of RCA and son of the founder of the company, took a trip around the world to visit RCA's foreign plants. He had recently announced the reassignments of some of the senior executives; also, the company had suffered alarming financial reverses from the 1973-75 recession and Sarnoff'sstrategic decision to enter the computer industry. In his absence, senior executives convinced RCA's board that Sarnoff's reorganizationwas inappropriate and that he should be replaced. It is reported that when he returned he was handed a letter of resignationfor his signature ("His Master's New Voice," 1975). The phrase that might characterizethis bloodless change in leadershipis "organizational coup d'etat." In a similar vein, in the late 1950s Brigadier GeneralHutton (faced with official opposition based on written agreements among the Department of Defense, the army, and the air force) developed the armed helicopter. Starting with baling wire and lashing machine guns to the frame, a group of midrank officersevolved the quick-strike,mobile air cavalry. The army's air force is now the third largest in the world, behind the United States Air Force and that of the Soviet Union (Bergerson 1976). The term that best describes the action of the Hutton group is "bureaucraticinsurgency." Finally, there have been many protest movements and rebellions within organizations. Fights between local and national unions with management and the national union on one side and the local on the other (Berg 1962; Serrin 1973), the recent cleavage within the Board of the New York Stock Exchange leading to the ouster of the president ("The Big Board's New Mr. Big" 1976), and organizationalschisms within the Southern Methodist groups (Wood and Zald 1966) can all be characterizedas mass movements in organizations. These phenomena are important for sociological analysis because they affect the major priorities of organizations, the control of organizational 824
825
American Journal of Sociology analogy. (The differencesbetween nation-states and organizations and the limits to the analogy are discussed in the Conclusion.) In this section we show first that a body of literature has been developing which implicitly or explicitlv treats social movements in organizationsin terms similar to those used in analyzing nation-states. Then we comment on the approach to social movements used here. Finally we articulate the main dimensions of organizationalpolitics and polities relevant to our analysis. Previous Literature Although they have not conceptualized them in social movement terms, a number of writers have described events in organizations that parallel descriptions of social movements in nation-states. There have been many studies of social movements within a particular organizational setting. In educationalorganizations,Heirich (1971) examinedprocessesof polarization and the spiral of conflict at Berkeley, Kelman (1970) focused on events at Harvard, Daniels and Kahn-hut (1970) reported on social movements and faculty involvement at San Francisco State, and Lipset and Altbach (1969) looked at the university student in politics and in university politics in different nations. In industrial settings, Britt and Galle (1972) extended Kerr and Siegel's (1954) classic study of strike proneness to show how the volume of conflict varies with proneness to conflict, extensity, and intensity of conflict. In other contexts, Martin (1954) and Wicker (1975) described riots in prisons which can be seen as protest movements; and Adams (1970) examined ideology, conflict, and heresy trials in Protestant seminaries. Finally, there is an increasing literature on change, ideology, and forms of protest and insurgencyin the professions (Needleman and Needleman 1974; Ross 1975; Epstein 1970; Gerstl and Jacobs 1976). This literature will eventually provide a rich base for secondary analysis. But from our point of view it is limited by its failure to set the problem of unconventional opposition and conflict in a comparative and more middle-rangetheoretical perspective. In contrast, we have identified six articles which begin to grapple with the problems of conceptualizing social movement phenomena in organizational settings. These articles use terms and modes of analysis very similar to those employed by analysts of political process, conflict, and social movements in the larger society. In a prescient and almost forgotten article, Selznick (1943) examined the battles for control of voluntary organizations such as political parties and trade unions. Although most of the article deals with the general theory of bureaucracy,the final section examines the relationshipof the rank and file to leaders, the growth of factions, and the ideologies of centralization and autonomy. 826
American Journal of Sociology mutinies and 20 strikes. He distinguishes three types of protest movements in organizations-promotion of interests, seizure of power, and secessionand argues that the form of the movements varied with the compliance structures of the organizations (Etzioni 1961). Lammers's article discusses the determinants of the incidence, duration, and outcomes of protest movements in military and industrial organizations. Although we believe his emphasis on the relationship of compliance structure to the type of movement is too simple (see Zald and Jacobs, in press), his article is important not only because it is an early example of the use of quantitative comparative techniques in the study of organizational social movements but also because it shows how a deep analysis of the structure of these organizations enriches our understandingof social movement processes within them. These six articles are important to us for several reasons. First, they are explicitly framedin terms of conceptsparallel to those used in the description of politics, unconventional opposition, and social movements. Thus they confirm us in our belief that it is useful to make the analogy more explicit. Second, all of these articles provide rich illustrations of a number of the processes that must be accounted for. Yet from our point of view they are limited, because they do not carry the analogy far enough. That is, they either limit themselves to a narrow range of organizations (e.g., voluntary organizations in Selznick [1943], hierarchic organizations in Leeds [1964]), or they limit themselves to a narrow range of social movement phenomena (e.g., mass movements in Lammers [1969], bureaucratic insurgencies in Leeds). Our task is more encompassing:it is to set the determinants of the occurrenceand outcomes of the broad range of types of social movements in the full range of organizationalsettings. Perspectives on Social Movements Since our approachdraws consciously on analogical techniques, the description of social movements in organizations employs the descriptive dimensions of social movements in society. A social movement is the expressionof a preference for change among members of a society. Social movements differ in the breadth of supportwithin the society (people who believe in the goals of the movement, contributors of time and money, activists and/or cadre), in their goals (the extent of change requiredand the level of concrete specification of goals), in the strategyand tactics of their carrier organizations, and in the locationof their supportersin the social structure. (These dimensions are elaboratedbelow.) There is disagreement among sociologists about the appropriateor most important analytic mechanisms for explaining the growth, decline, and change of social movements. Some sociologists emphasize the distribution and intensity of grievances and deprivation; others emphasize the role of 828
831
American Journal of Sociology military to resign; an officerwith 20 years of service can retire at half pay. All these are examples of situational determinants which affect the choice of organizationalmembers to engage in social movement action or to leave the organization. They are situational determinants of the forms in which discontent is expressed. Given the range of polity forms, the extent to which society may or may not impose constraints upon particular types of organizationsthat facilitate or hinder internal social movements and the microecology of groups and individuals, it is beyond the scope of this paper to offer a general theory of social movements in organizations.The variety of social movements and the range of polity situations are just too great to be fully analyzed in a single essay. Instead, we devote most of our comments to social movements in corporate hierarchical organizations. We discuss movements in federated and other organizations only briefly, to indicate the direction a fuller analysis would take.
SOCIAL MOVEMENTS IN CORPORATE HIERARCHICAL ORGANIZATIONS
Corporate organizations are those in which the units and facilities are "owned"by the group legitimated as the corporateoffice.That is, legitimate authority residesin the center (e.g., with board of directorsor chief executive officers).Among themselves corporateorganizationsmay vary in the extent of decentralization-centralizationof this legitimate authority. Federated organizations,on the other hand, are those types in which the units (locals, departments,chapters, or partners) have clearproperty rights and discretion which is established in the constitution of the focal organization (and possibly backed up by force of law). Moreover, they may have legitimate rights in the selection of executives and the establishment of policy. Voluntary membership associations, in contrast, can be either corporate or federated, but their distinguishing characteristic is that the members contribute resources upon which the central authorities depend. Thus authorities of voluntary associations have less discretion vis-a-vis organizational policy and are often in a precariousposition vis-a-vis members (Clark and Wilson 1961). Federated and voluntary organizations might therefore be called "upside down" organizations, for the flow of dependence is inverted from our conception of hierarchicallyarrangedorganizations. All of these organizationsmay experiencesocial movements. Hierarchical organizations differ from others in two major regards which have consequences for their political life. First, concentration of authority and power in a hierarchic structure means that there are higher costs of dissent and departures from organizational policy. Second, in such organizations subordinates have less normal access to the choices of major policy and of successors to executive offices. Conversely, in federated and voluntary 832
833
American Journal of Sociology though less proximal goals of policy and change are implied in the political action. Organizationalcoups can be contrasted with expected replacements (such as limited terms in office or retirement) and sudden successions which result from death. In the former, the succession is expected and conducted with a high degree of public awareness. In the latter, the succession is unexpected, yet once the initial shock of death wears off, replacementactivities follow the regular institutionalized procedures. Organizational coups, on the other hand, are unexpected and deviate substantially from routinized procedures.That is to say, they are planned and executed without the CEO's knowledgeor public awareness;otherwisethe CEOwould be able to retaliate. In addition, they involve the quick appointment of a new successor, which is at variance with the more usual procedureof search, trial, interviews, and selection. Five examples of organizationalcoups have been uncovered; here are brief illustrations of the climactic events in three of them. For MarionHarper,Jr., the sky fell shortlyafter 10:00A.M. [Interpublic:] on Thursday,November9, 1967. The setting was the windowless boardroomof the Interpublic Group,the world'slargestadvertising business,on floorof Manhattan's Time-Lifebuilding.The occasionwas the forty-fourth boardof directors a specialmeetingof Interpublic's with Harperpresiding as chairman and chief executiveofficer.... ... For nineteenyears Harper,now fifty-one,had been in chargeof the His energy,his ideas, his ambitionshad pushed, burgeoning organization. into a $700 millionbusiness.... bulledInterpublic dragged, and distributed in advance.... A twelvepage agendahad beenprepared WhenHarperbroughtthe first item beforethe board,one of the directors to question andrequested interrupted permission Taggart,the chieffinancial officer,on the group'scurrentfinancialsituation.... Harper,suspecting nothing,allowedthe question.... Anotherdirectorspoke.Taking note of the gravityof the situationhe movedthat the firstorderof businessbe the replacementof the chief executive officer.The motion was immediately The vote: six ayes, one abstention.[Wise1968,p. 136] seconded. of equalparts It was a pressconference MotorCompany] [Ford compounded of vinegarand butter. The scene was a Detroit gatheringcalled by auto that the FordMotorCompany mogulHenryFordII last weekto announce board of directorshad voted to relieve PresidentSemon E. Knudsenof his duties. The bitter taste stemmedfrom Knudsen'scalling of his own press conferencean hour and a half earlierto say he was "puzzledby this sudden Fordwas maneuand unexpected action."Beaten to the punch,Chairman in such situationsto "a mutualunderreferences veredout of the standard and was forcedto fall back on . . . "Sometimes these things just standing" don'tworkout." The butter, on the otherhand,was what wouldn'tmelt in the mouth of ambitiousLido A. Iacocca,whosesatisfiedsmilewas Knudsen's arch-rival, in sharp contrast to the sober expressionson the other Ford executives the PalaceRevolt at Ford"1969,p. 138] present.["Behind 834
American Journal of Sociology decent records of profit performance are not immune if they have created enemies. We have found several cases in which interpersonalhostilities and power battles led to the coup. For instance, if the CEO has recently taken steps leading to the demotion or lowered power of subalterns, it is likely that he has incurred their enmity. Until the demotion is fixed and they are isolated from power, the potential for a coup attempt is very high. Processes of interaction.-Regardless of the sources of their grievances, once a conspiratorialgroup has agreed to attempt to force out an executive, it must gain access to key board membersor stockholdersto carry the battle. Two authority processes must be considered. First, the CEO must be neutralized.Otherwisehe can counterattackeither by mounting a persuasive argument or by isolating and dismissing the conspirators. Second, board members and stockholders must be convinced. They must see the high cost of maintaining the current CEO arrangement.In two of the cases we have examined, board members had been contacted at earlier times but took no action then. The second time around, a more persuasive case was made. Moreover, it is clear that the CEOs were always surprised by the action and were often out of touch with their offices. In the RCA case described above, Sarnoff was visiting plants in Australia when the conspiracy took place. In the Genesco case, Franklin Jarman was on his honeymoon in Jamaica. In other cases a coup attempt may occur when the CEO is sick, and a sick CEO may recognize that a force out is coming. For instance, recently Donald Kircherwas replacedas head of the SingerManufacturingCompany (Hough 1975). He had warned subalterns and the board a year or two earlier that any attempt to use an interim management arrangement to replace him while he was sick would be fought. But when he became sick again and realized that he would not be able to resume the mantle, he easily gave up the office.4 Outcomes.-What are the outcomes of a coup? At least two types must be distinguished: the consequences or results for the participants and the consequences for the structural operation of the organization. The coup attempt may fail, as when conspiratorsmarshal support but board members dismiss the attempt as self-serving for the conspirators, and the CEO isolates or dismisses them. If successful, the coup may lead to a change in personnel, a change in the system, but have few other consequences, little change of the system. Students of coups in nation-states have remarked that coups often have little impact upon the larger society because they are not related to any underlying structural change. Coups in organizations
4CEOs who are forced out often receive remunerative severance contracts to erase their
sorrow: appointment as consultants with no duties and pay close to what they received while in office are common.
836
Bureaucratic Insurgency Bureaucratic insurgency differs from a coup in its target: its aim is not to replace the chief executive but to change some aspect of organizational function. It differs from a mass movement in extent of support and number of adherents. It resembles a coup in that for much of its duration it may be 837
Social Movements in Organizations development insurgency, the insurgents accept the overall structure of authority in goal setting but attempt to introduce new techniques for accomplishinggoals or refinementsof organizationalprograms.Middle-level officials or line professionals with discretion in allocating organizational resources, the insurgents pursue their own concept of organizational programs or product development while watching over their shoulder for elite interference. Social workers running group programs or allocating welfare funds, army officersdeveloping the armed helicopter, engineers developing the air-cooled engine at General Motors, or HUD bureaucrats attempting to push cities to have more racially integrated housing may all operate and be involved in this first type of bureaucraticinsurgency. Whistle blowing is a form of insurgency in which an insurgent deviates from loyalty norms to describe the disjunction between organizational functioning and public expectations. Whistle blowing may require only one insurgent. For instance, A. Ernest Fitzgerald, a systems analyst for the Defense Department, testified to the Senate on the massive cost overruns in building the Lockheed C-A5 cargo plane ("Defense: Ernest Fitzgerald RIF" 1969). Fitzgeraldacted by himself, but more organizedwhistle blowing is feasible. Undergroundnewspaperscan serve as outlets for private information that will discredit the elites. A third type of bureaucraticinsurgency focuses upon the main goals and policy choice of the organization.In 1949 the "admirals'revolt" questioned policies of the Defense Department that seemed to lead to a diminished role for the navy ("Revolt of the Admirals" 1949). Once the insurgency came out in the open, it began to resemble a mass movement. The difference between mass movement and insurgency lies in the greater openness of the former, the number of people involved, and the extent to which multiple organizationsor units are involved. The admirals' revolt was an insurgency at the top that was taken to higher authorities, Congress and the public. (The route xvas to Congressbecause the Defense Department was controlled by the executive.) One suspects that this form of large-scale and open insurgency at the top occurs only in corporate situations in which the insurgents have some degree of protection (e.g., good retirement plans or high job mobility) or in which the issue, the reward, the prize, is vital to the interests of the contesting elite. External support an(d authority response.-Whereas the organizational coup d'etat requires the involvement of key board members, a wider range of external supports is necessaryin insurgency.For instance, whistle blowing is fully dependent upon external support, for by definition organizational authorities have been unresponsive and typically have attempted to quash complaints. The insurgencymay also be aided by explicit material support. Army officers developing the armed helicopter cooperated w-ith small machine-tool companies which helped the officers machine and construct 839
Social Movements in Organizations organizationalgoals are multiple and units are only partially interdependent with each other. The formal recognitionof an insurgencyis similar to Leeds's (1964) process of "protest absorption,"but she addresseda much narrower range of phenomena. 4. Total incorporationof the insurgency depends upon the ability of the insurgents to get executive compliance or agreement. If over time the developed program or product can be shown to be consonant with the executive goals, the organization may fully incorporate insurgents' perspectives, and the insurgents may be promoted. Such total incorporation eliminates the raison d'etre of the insurgency. The adoption of the armed helicopter and related strategies of mobile infantry as a major component of warfare is an example of a totally incorporatedinsurgency. Three hypotheses summarizeour argument about insurgency: 4: Insurgencies are most likely to occurwhereprofessional and Hypothesis normativecommitments base for perspectiveson providean independent goals,products,and policies. Hypothesis 5: The morecomplexthe organization and the moredifficultthe the longerthe durationof insurgency. surveillance, 6: The outcomeof insurgency Hypothesis dependsuponthe extent to which the insurgents of suppression. threaten authorityand the costs to authorities Mass Movements: From Protest to Rebellion As noted earlier, bureaucratic insurgencies may span many levels in an organization. They range from small cabals at the top to the concrete and concerted efforts of middle-level managers and professionals. When small, they may have a conspiratorial cast. On the other hand, as they become more organized, they may develop coordinating committees or caucuses which begin to resemble mass-action movements. As their numbers grow and their tactics move toward withdrawal of labor, petitions, and boycotts, they begin to resemble mass movements. Mass movements at the nation-state level range from movements of protest to rebellion. They are defined as collective attempts to express grievances and discontent and/or to promote or resist change. They vary in goals from those aimed at melioristic change of specificpractices or rights to those aimed at redefining the distribution of power, constitution of rules, and norms of society. By analogy, mass movements in organizations range from the expression of minor grievances (not previously acted upon by authorities) to major attempts to seize control of the organization. Mass movements differ from insurgenciesand coups in number of participants and the visibility of their actions. Coups and insurgencies may be restricted to face-to-face groups and become visible only toward the end of their histories. Mass movements, on the other hand, may be initiated by a 841
American Journal of Sociology small group, but larger numbers must be mobilized if the initiators are to gain their ends. Examples of mass movements are work slowdowns, wildcat strikes, mutinies and secessions, mass desertions, and prison riots. It is important to note that, while the proximal goal of the movement is to change the behavior and goals of organizational authorities and the structureof organizations,the real goal may be changes in the largersociety. Thus the student movement disrupted the University of California at Berkeley, Columbia,Harvard, Michigan, and Wisconsinand had as proximal targets changesin administrativebehavior. But the real goals were to change the behavior of Lyndon Johnson and Richard Nixon or, in more radical form, the structure of American society. Similarly, a plant seizure or strike may occur to achieve concrete gains, but it also may be a weapon for changing governmentalpolicies or the government itself. A paradox confronts us in thinking about mass movements in organizations. In discussing coups and insurgency, we drew on sparse evidence; on the other hand, when one turns to the study of mass movements, the literature appears rich, systematic, and quantitative. After all, economists and sociologists have been studying the factors related to strikes, industrial conflict, and unionization for two generations. And more recently quite sophisticated quantitative analyses have decomposed the determinants of the numbers, breadth, and duration of strike activity (Britt and Galle 1972). Yet the first glance is deceiving. For strikes need not represent the social movement-like phenomenon of unconventional politics. They may representa fully institutionalized aspect of the collective-bargaining process. As such they are part of the normal and institutionalized political system of organizations. Winning union recognition and the legitimation of strikes is part of a social movement process in society and in specific organizations. Yet the union has now become one of the political mechanisms for aggregating preferences and resolving conflicts in the organization. Snyder (1975) has shown that one can predict aggregate strike behavior from aggregate economic variables after a collective bargaining system has been institutionalized. From our point of view, institutionalized unions that engage in strike action are part of normal politics. Of course there are some parallels between industrial strikes and mass movements in organizations. Both involve mobilization of workers and the calculation of costs and benefits. Yet the mechanisms for mobilization, the costs of organizingfor collective action, and the extent of organization and societal support are different enough so that we cannot incorporatewholesale the literature on industrial conflict in our explanation of organizational mass movements.5
5The problem here parallels an analytic gap in the study of social movements and politics in society. Party politics and the aggregation of preferences through the institutionalized 842
843
American Journal of Sociology within a subordinate group and the possibility of alliances between subclasses of subordinates and superordinates. This hypothesis follows Kerr and Siegel (1954). Thus we would expect that organizations with little occupational differentiation would be more likely to have superordinatesubordinate conflicts. Hypothesis 9: The greaterthe vertical segmentation, the morelikely that grievance channels, mobility channels, and communicationchannels in generalare blocked. This hypothesis has been proposed in the social movement literature (see Oberschall1973) and offeredas an explanationfor strike proneness (see Kerr and Seigel 1954). Of course it dates back at least to Pareto. In organizations in which occupational status and ethnic class status are overlaid, we would expect segmentation to be especially strong and both within-groupsolidarity and the chances for resentment to be high. 10: The moredifficultor costly it is to exit and the greaterthe Hypothesis commitmentto the incentivesof the organization, the more likely social movements are to form. In some ways this hypothesis, which just restates Hirschman's (1970) exit-voice thesis, parallels Thibaut and Kelley's (1959) analysis of "comparison-level alternatives." Thibaut and Kelley, of course, were interested in whether people continued their line of action within a social interchange or moved outside that interchange. It is apparent that different personnel in organizationsand positions are confrontedwith differencesin comparisonlevel alternatives. The value of the comparison-level alternatives can be broken into two components-the cost of exit and the value, negative or positive, of the perceived option once attained. Comparisonlevel is the net value of the current situation or line of action minus the cost of exit and the value of the goal, once obtained. Some organizationshave extremely costly exits. Military institutions raise the costs of exit very high. Nevertheless, it is apparent that exit is less costly in the army than in the navy at sea. Under relatively similar levels of disaffection and grievance, we would expect soldiers to mix use of desertion and protest, whereas sailors at sea use only mutiny. Blake's (1976) study of military resistance to the Vietnam war reports cases of mutinies in military prisons and among troops on the line (refusals to move into combat), but it is our impression that the number of actual mutinies on the line is much lower than in the more contained prison situation. All of the mutinies that Lammers (1969) studied were naval ones. There are, of course, cases of whole units of armies deserting; these should probably be seen as secessions. They are not protests of the bottom against the top; instead, they are led by dissident officerswho disagreewith either the treatment of their units by the authorities or the policies of 844
Social Movements in Organizations authorities in prosecuting the war and in relation to subject populations (see Solzhenitsyn 1973). Earlier we noted that strike rates do not necessarily reflect social movements in organizations; they do, however, reflect disagreements with management offers or existing conditions. With regard to the hypothesis under consideration,Stern (n.d.) finds a negative correlation,over communities, between quit rates and strike rates. A final derivation from the exit-voice hypothesis can be suggested. The classical situation for exits is free markets with low costs for information and transportation. There buyers or sellers easily compare options, and consumersswitch from seller to sellerwith no protest. However, our hypothesis suggests that consumer protests increase as monopoly increases-a relationship amply confirmedby American history-and monopoly can be defined as the opposite of exit choice. Monopoly exists when a single seller controls the sale of a good and there are no close substitutes for it. The exit-voice hypothesis has a strategic role in accounting for the structural conditions of mass movements in organizations. First, it allows us to account for variation within an industry (defined as a set of organizations offering somewhat similar products) of two forms of expressing dissatisfaction, exit and voice. Second, it helps to provide an answer to Olson's question: Why collective action, when collective action is accompaniedby free riders?The implication of Hirschman's argument is that, if one option for individuals, exit, is removed, an individual must weigh taking no action against taking some action that will promise a surplus of benefits. Stated differently, the removal of the exit option raises the comparativediscounted benefits of the voice option. But Olson's challenge still remains. Our next hypothesis treats it more directly. 11: The greater the associationaldensity and the higher the Hypothesis who are members of associations, of organizational proportion participants the easierit is to mobilize. A majorpart of the critique of the mass-society theory of social movement has been that people who are most involved in voluntary associations and political organizationsare also more likely to be involved in social movement action (see Oberschall1973). There is supporting evidence in organizations. Although not dealing with social movement participation, Lipset, Trow, and Coleman's (1956) study of the International Typographical Union makes it clear that the associationally dense culture of the ITU facilitated political involvement. Similarly, studies of the student protests find that antiwar protest was correlated with the proportion of students in left organizations (Peterson 1968). And Von Eschen, Kirk, and Pinard (1971) found that early participants in the civil-rights-movement sit-ins were likely to be part of established, politically oriented organizations, often 845
846
AmericanJournal of Sociology moves are to ask for the involvement of political authorities. Similarly, in many countries (most Latin American countries, France, etc.) the universities are not autonomous of the ministry of education. A mass movement in the university, especially if it moves into the streets, is responded to by political administrators, not university officials. Where officials do have discretion, several strategic policies may be adapted. Lammers (in press), in a comparative study of the response of university authorities to student movements, lists four general combinations of strategies, tactics, and goals: Strategies Repressive Concessive Preventive Experimental Tactics Fight off Buy off Standoff Join in Goals Elimination Appeasement and Nonemergence dissolution Copartnership
In the first, the authorities attempt to eliminate the movement and its leaders. They suspend students and use heavy sanctions. In the fourth, the authorities appear to buy in to the goals of the movement and move to align the organization with the goals of the movement. Lammers goes on to assess how differenttypes of tactics will work in differenttypes of universities. For instance, he is skeptical about the viability of the experimental strategy and the tactics of joining in in large universities, and he believes that preventive strategy allows universities to continue to carry out their educational and researchmissions with less disruption than the use of either concessive or repressivestrategies. (He does not considerin detail the interaction of opposition and authority strategies.) Outcomes.-In mass movements, immediate outcomes depend upon the ability of partisans and authorities to sustain conflict and the extent to which there are viable options for bargaining. If situations are defined in zero-sum terms, a sheer power calculus can be used. But in open-conflict situations, a variety of symbolic and partial solutions may permit both parties to win. Mass movements with specific and narrowgoals may accomplish a change in the system of a specific organization, with little impact upon society. On the other hand, movements with broad goals are likely to fail unless they are part of a broadermovement in the society at large. The ability of partisans and authorities to maintain a conflict depends on their ability to mobilize resources and on the continuous availability of resources. This is especially important in understanding the difference between mass movements in such organizations as universities, with high turnoversof personnel,and those occurringin such organizationsas factories, with greater continuity of personnel and organization. Thus, although in a university the movement may be at high tide in the spring, the lack of continuity of student generations and the departure of student leaders at the end of their senior year lead to a high cost of remobilizingresources.On 848
4-) 0
rn Cd tD
> 0.) -0
Cd
1-4
Cd
4-)
o
r. Cd
4-)
Cd to
0
N 4-J
Cld o cd w
4-J 4-J o o
1 4-1
to Cd s.
w
(L) u
'O C)
W
0
Cd
4-)
El)
-0
0
En
>
0.)
El)
cn
4-)
-4.J 4-)
0 Cd
C 1.4
0 w
Cd
rn
0 u En
-4
4-3 0 o >-. -
to.-
4-)
Cd
Cd
cd w Cd-40 cn
0 41
4-J
b,O b,O ce
4-J En cn 1-4 En w o
w
4-) U
(L)
4-) 1W cd W (L) ,8 0
cd
4-J cd
11-4
-,O
w cd
4)
Cd
4
to 0 4-).
>
En
cd N4 Cd ho
;t
El)
cd
4.j Cd
En U)
W 4-) W 0
ho 0 cd W ci
>.-. 0.)0 4-)
tn 0
>
4-J _-q
Cd
u
14-4
u Cd Cd u C) .- -ci
14-4
0
w 4-1
lc C).O r. ;$ En Cd
Cd
ho
>
0 0
$. U) z
Cd Cd
cd
's.-4
Y)
4-J
(L)
En
4-J
U)
cd
4-)
El) O
bD 0
En
bo 0 cd
cd ho
z
u 0
a) 1-4
Cd
i -4
En
Cd 4. Cd
>
cd O bo
C-)
Cd 0 0
cd
4-3 0
0
C/)
bjDEn Cd 0 bD Cd En'O Cd U) u
W 0 :3 cl 0
r4-)
tn 0 z
0 0
Cd 0 bD Cd w
0 4-)
>1
C; r.
lc
1-4
bO
r.
0
Cd 0 ;:1 -0
4 Cld
Co
0 >,. C) Cd
a)
w 0 (L) u > -1 0 Cd
4-) $-4 Cd
w
4-)
1-4
E,11 4
Cd
Cd ho
cd
W
;.
0 0 cd
a)
0
En 4-)
En
Cd
>1 1-4
a) cd > tn 0 '-w
cn 0
-4
0
En 0
4-)
4-)
Cd
C/) C/)
0 -C C)
Cd
L4-4
cd U 4-) bjD,--, .0 0 tn
1-4
CdN -4 Cd bo,,
r. : -= m
0 Cd
bD q
x.,
ri)
4-)
blo
C)
4-J
(U
4-)
4-)
Cd
4-) 4-J
4-)
Cd
En 4-) 4-J
>
4-)
Cd
En 4-) rn 4-)
0 ho
4-)
Cd
Cld 0
.0 Cd
->
4-J
Cd N
0
Cd
(-
American Journal of Sociology Locationin the social structure.-Social movements differ in their location along vertical, horizontal, and sociodemographicdimensions. The vertical dimension refers to the location of adherents in the hierarchicalsystem of the organization. The horizontal dimension includes both the spatial and functional differentiationof the organization.These differentiationsprovide the basis for the development of within-group solidarity and differentiated concepts of organizational mission, problems, and priorities. The sociodemographic dimension refers to the age-sex-social background of groups that combine to shape orientationsand perspectives within the organization. In an organizationalcoup, the palace guard is located close to the authorities. That is, the guard actually will be the elite in the status hierarchy and therefore experience very little spatial or functional differentiation. This lack of differentiation should lead to high within-group solidarity and congruencewith the board concerning organization mission, problems, and priorities. Such congruence is bolstered by similar social backgrounds,age, and sex. In contrast, the orientation of the insurgent department is likely to diverge from that of the authorities and be predicated on the distance from them, on spatial and functional differentiationvis-a-vis other parts of the system, and on wider variations in age, sex, and social background. Finally, rebels will probably be most detached in their orientation from authorities due to high vertical separation, extreme spatial and functional differentiation,and wide variations in age, sex, and social background. Linkage with externalelements.-Coups require linkages with key stockholders, outside board members, or key financial supporters. Insurgencies, on the other hand, are fueled by ideological and professionalsupport. Mass movements with broad goals are heavily dependent upon support from outside the organization.These include the cooperationof agents of control (police), material resources,and political and ideological support. Duration.-Finally, the duration of conflicts varies considerably. The coup, though possibly brewing for a long period, is over in a few short hours; mass movement and insurgency, on the other hand, can continue for long periods of time. In every case we have been consideringin this section, there is a central legitimate authority and a relatively clear hierarchicdistribution of power. We now describe social movements in organizations in which the central authority is usually weaker vis-a-vis the units or members.
SOCIAL MOVEMENTS IN FEDERATED AND OTHER ORGANIZATIONS
A complete analysis of social movements in organizations will include descriptions of the variety in social movements across the wide range of organizations that diverge on major constitutional dimensions, such as voluntary associations of different types (e.g., organizationswith purposive 852
Social Movements in Organizations goals such as political parties, social movement organizations, specialpurpose interest groups, professional associations, associations conveying solidary incentives). These organizations differ from corporate hierarchical organizations in several regards relevant to social movement and political processes. First, members and units typically stand in a different relation to the organization and to the center, the titular authorities. The vertical stratification system does not dominate the system of rule. Most voluntary associations do not control major stocks of incentives of the members; thus exit is easier. Time commitments of members are low, and the authorities are less in a position to impose strong sanctions. Second, the constitution of authority in these organizations restricts the rights of the center vis-a-vis membersor units. In federations, ranging from business partnerships to labor unions to Protestant churches, the units are basic elements of the organization,retain rights of operationin their own domains, and share in decisions about the direction of the total organization. These organizations possess political systems which, if not democratic, at least establish a flow of authority and legitimation from the members and units to the central organization. They establish procedures for succession to major office that involve the unit or members;they establish proceduresfor the membersor units to affect majorpolicies and directionsfor the organization. What effect do these differencesin polities have on the kinds of social movements that emerge? As noted earlier, professional associations, voluntary associations, and federated organizations have more open politics than do hierarchicalorganizations. At least dissent, factionalism, and conflict are less likely to be repressed. In many ways, then, one can move the whole apparatus of political sociology analysis into these polities with even less modification. Moreover, theoretical and empirical works have examined phenomena resembling social movements in these organizations before, though often under different guise. Thus the literature on factionalism and democracy in unions (see Edelstein and Warner 1975), on the conflict in the Methodist church (Wood and Zald 1966), on factionalism in political parties (Wilson 1966), and on change in professions (Ross 1975; Epstein 1970) begins to describe the phenomenon in question. In general we expect that coups and insurgencies are less likely to occur in these organizations,and factional fights and secessions are more probable. Federations A federated organization is defined constitutionally by the rights reserved to the units. Both constitutionally and in practice, there is variation in the relationship of federated units to their centers and to each other (Sills 1957). For instance, a federated union may have weak and apathetic units, 853
AmericanJournal of Sociology so that the central organization essentially dictates policy and makes it extraordinarilydifficult for any local initiative to materialize. In contrast, other units may command resources, loyalty, and a spirit of independence that make them a perennial thorn in the side of both management and central union offices (Kuhn 1962). Similarly, church organizations may range from episcopal structures in which the living constitution gives great power and command to the bishop to congregational structures in which the local congregationcontrols and appoints the minister, owns the church, and does not need to accept any policy directives from the center. Social movements in federated organizationsusually take the form of rebellions of single units or coalitions of units against the center. Conspiracies may occur if the dissident faction's compositionor goals are abhorrentto the dominant coalition in the organization. For example, the establishment of front organizationsby the Communistparty is not such ancient history, nor is the conspiratorialtakeover of the Social Democratic party by Lenin-led Bolsheviks. The greater the degree to which local units control (own) the resourcesof the units (the physical facilities and/or the loyalties and identification of the members), the more likely it is that schisms and secession will occur. Thus, as compared with hierarchic corporate organizations, social movements in federated organizationsmay lead to a breakup or splintering off of various segments of the organization.The history of AmericanProtestantism is rife with such schisms and secessions. Schisms also occur in political party systems where proportional representation rules exist. In contrast, when voting rules lead to winner take all, political parties retain their factions. There are essentially two types of schisms. In one, whole units secede; in the other, groups of members within units find the program or policy of the organization unacceptable and secede with the intent of setting up a new organization. The former is probably more easily accomplished, since the whole unit secedes with an organization and facilities already in place.7 Less dramatic than schisms or secessions, members or units may lead the battle for changes in policy. For instance, the Episcopal church, which combines hierarchical and federated features, has had a long, drawn-out battle over the adoption of a new translation of its book of prayer. This movement has been led by college English professorsand has brought about a reversal in church policy. By the same token, trade union locals may hold out against the settlement negotiated by the international union. This is not to suggest the center is without power in federated organizations. For instance, a recalcitrant local can be put in trusteeship and its officers,
I In episcopal structures the church building is often owned by the central church, even though possibly paid for by local donations. And in court battles over expulsion, the building may revert to the overall association. On the other hand, in congregational structures the local denomination may own the building.
854
Social Movements in Organizations stewards, and business agents suspended. Since putting a local into trusteeship also leads to cutting off funds and abrogating votes, it is a potent weapon for the center. In Protestant denominations, the center's power over the dissident congregation ranges from censure to lockouts and even to heresy trials. Voluntary Associations Social movement organizations, professional associations, social clubs, and local community-bettermentgroups are characterizedby heavy dependence upon their membershipand membershipcommitment for their maintenance. As noted above, because the organization is not a purveyor of major stocks of incentives, the center must curry favor; and executives are heavily dependent upon persuasion, symbolic rewards, and providing services and friendship to maintain commitment. In some voluntary associations, open conflict and politics are muted precisely because the conflict would lowTer the commitment of members and lead to high rates of exit. Organizationsthat depend upon solidary incentives attempt to avoid conflict; they develop a politics of accommodation. If conflicts do emerge, they may lead quickly to individual or group secession. Even voluntary associations with purposive goals must avoid conflict, for the losing factions are likely to secede. Professional segments.-One more type of social movement in organizations deserves mention: a movement in a profession that infiltrates both corporate hierarchical and professional associations. Bucher and Strauss (1961) describe the growth of segments in social movement terms. The members develop a sense of group identity and difference from the rest of the profession; they develop a program of action. In professional associations, such segments operate loosely either through caucuses or through attempts to change the policies and structureof the association. In hierarchic organizations, they can be described as bureaucraticinsurgents.
THE UTILITYOFTHE MODELAND RESEARCH PROSPECTS CONCLUSION:
This analysis has suggested that the similarities of nations to organizations permit us to utilize concepts of social movements drawn from the former to examine similar processes in the latter. Much more than a metaphor is intended. Organizations and societies are similar enough so that a strong analogical model can be and has been employed. There are indeed strong similarities.Both organizationsand nation-states, for example, are composed of groups located in a stratified system of statuses and with differential access to resources.They are both webbed by ideologies which often cause conflict in terms of which perspective will lead more efficiently to goal 855
American Journal of Sociology accomplishment and whose perspectives on goals ought to be adhered to. They both have group cleavages. They both have systems of rules and of social control, and they both have group and individual resistance to their rules and control. On the other hand, there are differences that weaken the analogy. A state has a legitimate monopoly over coercion, whereas an organization does not. Moreover, a state has a pluralistic set of goals. It is not even clear how one can conceive of the goals of a state; whereasthe goals of an organization seem at least more concrete. Nation-states are relatively large in territory, whereas organizations are more concentrated geographically. Nation-states can present great obstacles to mobility (i.e., entry and exit), whereas in most organizations, except for coercive types, entry and exit are comparatively easy. Finally, the cleavages and structure of nation-states endure for generations and are transmitted through families and the class system, while the cleavages of organizationshave a less enduring base. The question then becomes, What are the consequences of these differences for the strength of the analogicalmodel? First, the claim that the state has legal powers of coercion, whereas the organization does not, suggests that organizations will not have a full range of strategies with which to deal with conflict groups. Yet it can be argued that the organization'sability to expel, isolate, or apply sanctions to its members represents similarities to legitimate coercion and that both nation-state and organization use coercion which is circumscribedby legitimate norms. Second, the claim that goals in a nation-state are more abstract and organizational goals more concrete implies that movements may be easier to mobilize in an organization.Yet it can be argued that in both cases there is a duality of goals. That is, there is a distinction between official and operative goals. Officialgoals are the entities' general purposes, as set forth in its charter, whereas operative goals are the actual goals sought which are determined by group interests (Perrow 1961). Furthermore, it can be asserted that the recent civil rights, end-the-war, women's, and environmental-protectionmovements have all had very precise goals. Third, the argument that states are larger and organizations smaller implies that mobilization costs may vary as a function of size. In some sense this may be true, yet it can also be argued that the processes are the same and only the scale is different. If both entities were standardized to control for size, the results might be very similar. Fourth, the assertion that the two entities differ in mobility potential implies that the exit option may be readily available in organizations, hence leading to fewer movements, whereas exit from one's nation-state is more difficult, suggesting a greater reliance on voice (i.e., movements). Yet this reasoning ignores the facts that (1) changing one's job represents a serious disruption; that is, while job mobility may be theoretically possible, on a practical level it is not 856
857
American Journal of Sociology movement is like looking for a needle in a haystack. (See the critique and programmatic statement on this problem by Snyder and Kelly, in press.) At least in organizations the boundaries of effects have clearer space-time limitations. Specific policies, authority relations, and changes in rights and prerogatives can be pinpointed. This article has identified a new agenda for organizational analysis. Instead of following the traditional approach to conflict, power, and exchange, the essay conceives of many such processes as unconventional politics that give rise to movement-like phenomena. More specifically, the essay distinguishes organizational coups, bureaucratic insurgencies, and mass movements to argue that a combined approachof analogical theorizing and resource mobilization may explain the form, occurrence,and outcomes of such phenomena. After offering a number of propositions relating to a resource-mobilization perspective, we evaluated the strength of the analogy and implications for research. The model we have presented here has, we believe, several uses. First, it brings together under one conceptual structure partial analyses of social movement phenomena employed by others. That is, we are able to link a number of analyses of insurgency, strikes, enclave formation, and other phenomena that have been conducted without reference to each other. Second, our approach suggests a number of possible research strategies to illuminate a phase of organizational life that has been hidden in darkness. Third, it suggests a new paradigm in the study of organizations:the study of mobilizationprocesses (see Benson 1977). Finally and most provocatively, it opens up once more a host of questions that have seemed settled. Can we, for example, look at the Protestant Reformationnot as an issue in theology, or as an issue in the relationshipof societal constitution to collective ideation (Swanson 1967), or an issue in state formation but as an organizational rebellion?Seen from Rome, the Protestant reformersmust have looked like bureaucratic insurgents, and the insurgents had to mobilize internal and external support.
REFERENCES Adams, Robert L. 1970. "Conflict over Charges of Heresy in American Protestant Seminaries." Social Compass 17 (2): 243-60. "Behind the Palace Revolt at Ford." 1969. Bitsiness Week (September 20), pp 138-41. Benson, J. Kenneth. 1977. "Organizations: A Dialectical View." Administrative Science Quarterly22 (1): 1-21. Berg, Ivar. 1962. "The Nice Kind of Union Democracy." ColumbiaForum 5 (Spring): 1823. Berger, Michael. 1975. "Organization Coup d'Etat: The Unexpected Social Movement for Succession." Xeroxed. Nashville, Tenn.: Vanderbilt University, Department of Sociology. Bergerson, Frederick A. 1976. "The Army Gets an Airforce: The Tactics and Process of Insurgent Bureaucratic Politics." Ph.D dissertation, Vanderbilt University.
858
859
861